A fair amount of material have already been produced about theories of change (TOC), especially since the beginning of the 2010’s when its use started to spread out in the field of international development. So, I thought that a good way to start was to share some of the most relevant resources (methodological documents and videos for those with limited time availability). My purpose here is to help you find all the basic concepts in one place. And if you are more experienced with TOC, the following lines will certainly help you reflect on your experiences and anchor your learning. In each section, I start with a presentation of what I found. Then I add my two cents, using my experience in TOC design and trying to highlight what makes the difference between a TOC that looks good on paper and one that is actually useful in practice.
Nowadays, one of the first things you are asked about in a program or project design is: What is your theory of change? Indeed, it became an essential part of the common vocabulary in many development agencies. But beyond “speaking TOC”, you are probably expected to include it in your program design. So if you don’t feel 100% confident about the basic concepts or if you want to get insights from actual practice, enjoy the reading and please share your thoughts and comments afterwards.
1. TOC: what is it and what for?
What I found
For P. Rogers-UNICEF Innocenti (see all references below), a TOC is a description of the pathway that allows you to go from a current situation to an intended one, “and what needs to be done to move from one to the other”. It is also sometimes presented as a more detailed version of your vision.
Regarding purposes, three cases are often identified:
An institutional purpose, usually related to a mission statement
A programmatic purpose: explaining a program logic to internal and external stakeholders
An evaluation purpose: comparing the initial intent with the expost reality or actual changes (indeed, a large influence on TOC comes from the field of rigorous impact evaluation).
C. Valters (on the basis of the Asia Foundation's experience) found a useful and effective way to summarize TOC purpose: “communication [usually towards stakeholders], learning [as an internal good practice] and accountability [expectations v.s. achievements]”.
My 2 cents
In the first place, I would emphasize the word “theory”. In science, a theory is usually defined as a set of ideas and hypotheses based on observation, experience and inferences. But more important, a theory has an instrumental value: it (usually) exists for the sake of being tested and verified. Idem for a TOC. For instance, this morning I read a World Bank blogpost about teacher pay for performance (ref below). In theory, you would expect that giving a bonus to teachers based on their students’ learning improves overall performance. Well, available evidence is not as clear-cut as you would expect. Either you use it for evaluation or programmatic purpose, keep in mind that a TOC might also end up being proven right or wrong by facts. And that’s where one of the important values of TOC lies. The absence of a TOC would be like navigating without a clear and transparent intention: “I am going somewhere”.
Second, as we have seen, TOC are usually defined as a pathway, from a situation A to situation B and how to get there. This definition might give the sense of a linear process. The reality is more complex and here there is a clear tradeoff between describing our intention in simple terms and describing the complexity of multiple pathways. Actually, it is more useful to think about TOC as an ecosystem, rather than a journey and I’ll come back to it in the next section.
2. How it should be built?
What I found
Starting point. Across all material reviewed, there is a large consensus on the fact that a TOC should start from a solid analysis of the root causes of the current situation. In his short video, D. Hearle refers to the problem tree approach with a specific emphasis on the roots. The Chemonics video also provides a concrete example on how to perform the root cause analysis. And usually, they all refer explicitly or implicitly to the Five-Whys method developed by Toyota Production System in the 1950’s (Free tip: if you are doing this analysis in a workshop, a good exercise is to make groups of 5 individuals, assign them a problem, ask each of them to define “one Why” and ask them as a group to sequence the “Whys”). This method has also been further developed under the name of Fishbone or Ishikawa diagram.
Beyond the mere product, a TOC is a process. This is a point that Vogel highlights from her discussion with practitioners and it has a significant impact on how it should be done. In particular, it means more time dedicated to the process, internal and external stakeholders involved as early as possible (at the level of root cause analysis preferably), and managing and making the best out of multiple perspectives.
My 2 cents
Think about TOC as an ecosystem. In an ecosystem, when one part moves, the other parts move as well. Kind of a domino effect, but giving the ability to each domino to stand up and fall on the other side. With this representation in mind, you are able to think your TOC as a set of interactions, not only as a set of one-way and linear reactions. Then the changes that you are able to introduce in this ecosystem will end up in a series of other changes.
Picture TOC as an ecosystem and logframe as a production function (of each change). The distinction between TOC and logframe is still highly confusing for many, including in some of the materials presented here, mainly because both come from the same field of origin. Nothing better than a good example. Let’s imagine that we introduce a change in today’s Earth ecosystem. This change is the revival of a few dinosaurs. We will probably anticipate that there will be a few reactions and interactions in our ecosystem (if you lack imagination, you always get back to Jurassic Park). Here, the logframe would be the production function that leads to the dinosaurs’ revival, starting from the needed R&D resources, the research team recruitment, their research and outputs on dinosaurs’ DNA and the necessary incubation process, for instance. With this example, you can also have a sense of the power of one’s assumptions on the change interactions in your TOC. Assuming here that the resulting dinosaur will end up being herbivore or carnivore makes a huge difference.
To sum up, thinking in terms of ecosystem and production function has several advantages:
It forces us to acknowledge the complexity of the environment in which we are intending to introduce changes
It invites us to consider several perspectives (from multiple stakeholders) and understand the possible reactions and interactions produced by our intention
Personally, it also helps me making the distinction between what goes into the theory of change and what goes in the logframe (and this is a big weight out of my shoulders I have to say!)
It helps integrate the consideration of demand-side changes, not only the supply-side ones. Usually, TOC put a lot of emphasis on: “If we produce this, then this will happen…” and we tend to belittle all the behavioral changes that are needed to achieve the end goals.
Among others that you can share in the comment section
3. How a TOC looks like?
What I found
Vogel lists the the basic elements that make up the TOC approach: context, long-term change, process/sequence of change, assumptions, diagram and narrative summary. Others add to this: a clear identification of stakeholders, or even further a stakeholders’ analysis, partners of change, or higher-order global end goals, such as the SDGs, among others.
In practice, because of the above-mentioned TOC-logframe confusion, some (maybe most) of the TOC we find look like a logframe (some explicitly including inputs and outputs in their description for instance), and have a very linear aspect (ref: video from SoPact below, as well as Rogers). The advantage of such representation is that preference is given to clarity over complexity. So, if such a model is used, it should be clearly acknowledged that many reactions are not captured in the representation or should be extensively discussed in the narrative.
My 2 cents
The models that make their best effort at including more complexity in their representations and keeping a good balance of clarity are the ones that include two key elements: time and space. Indeed, all changes do not happen at the same time and in the same sequence. Usually, our intervention’s scope is limited to short-term changes, which are expected to have impacts on longer term changes. Or, better practice still, we might describe a sequence of short- and mid-term changes that we plan to act upon over a significant period of time (5-10 years for instance). Similarly, the intervention might target different levels, usually micro, meso and macro levels. Micro-level are changes usually at the level of the individuals, meso- at the level of intermediary entities (ex: decentralized administration, a technical direction within a Ministry, a village association, etc.) and macro- at the national level. Those distinctions are key to identify levers that are actually actionable at the level of our intervention, in a sequenced timeframe, defined spaces and related to both supply-side and demand-side factors.
Below is an example of such a representation:
From Vogel and originally from: United Nations Population Fund and United Nations Children’s Fund, Joint Evaluation, UNFPA–UNICEF Joint Programme on Female Genital Mutilation/Cutting: Accelerating Change, 2008–2012, Final Report, Volume II, UNFPA/UNICEF, New York, 2013, annex 13, p. 192
Last thoughts...
In conclusion, we can see that there is no one recipe to do a theory of change. In the end, the good one is the one that suits your purpose, whether you are intending to improve health outcomes in a given country or to revive dinosaurs on Earth. In my opinion, TOCs brings most of its value when learning and vision alignment happen during the process. It won’t come as a surprise to you that I would encourage you to make an attempt at embracing the complexity of your ecosystem, while keeping a usual dose of pragmatism, rather focusing on the quality of the end product.
Now, we have seen that most of the material comes mainly from the early 2010’s and not much is found in the 2nd part of the decade. There might be a case for updating the approach. Indeed, the approach is starting to be challenged in some organizations by what we call the “system mapping” approach, which comes from the design thinking and innovation fields. It brings a clear recognition of the complexity of the interactions within a given ecosystem. It also puts a lot more emphasis on the micro-level (people), non-linear interactions and their implications for the ecosystem as a whole (I include references below). It is coming… let’s keep an eye on it and get the most inspirations out of it.
What is YOUR experience with TOC? How do you see the future of TOC? Tell us in comments…
References
Documents
Rogers, P. (2014). Theory of Change, Methodological Briefs: Impact Evaluation 2, UNICEF Office of Research, Florence.
And the filmed presentation: https://www.youtube.com/watch?v=KRptX_DNL2Q&t=70s
Comments: the document is not very structured but there are some important pieces into it, such as the graphic representation that I used above.
Valters C. (2014). Theories of Change in International Development: Communication, Learning, or Accountability?, JSRP Paper 17
Comments: The overall architecture of thought is really effective; then the discussion is quite specific to some types of institutions such as DFID and the Asia Foundation.
Vogel I. (2012). Review of the use of ‘Theory of Change’ in international development. UK Department of International Development.
World Bank blogpost about Teacher Pay for Performance (2021) https://blogs.worldbank.org/education/teacher-pay-performance-does-it-really-work
Videos
Theory of Change Explained, David Hearle
Comment: the explainer for me at this point
Theory of Change: It's Easier Than You Think, by Chemonics International
Comment: not really comprehensive but a few simple, effective and graphic messages.
Theory of Change - Foundation for Social Impact Measurement, by SoPact
Comment: Interesting and structured effort, though a bit of confusion between TOC and logframe
Fun video about the 5 Whys
Comment: Other videos I found are quite boring, too long or not really grounded in solid knowledge. But if you find good ones, please share.
Blogposts
C. Alford, 2017. How systems mapping can help you build a better theory of chan
E. Gray. 2020. Systems Mapping: A Vital Ingredient for Successful Partnerships
Comments